docs: clarify draft bind input behavior

This commit is contained in:
lda
2026-06-29 19:36:04 +07:00 Verified
parent ca299e6fa0
commit af7452a794
6 changed files with 26 additions and 9 deletions
+5 -3
View File
@@ -404,11 +404,13 @@ def bind_draft(
typer.Option("--to", help="Target path, for example local.x or state.y."),
],
) -> None:
"""Bind a capability step path and project the matching schema.
"""Bind a capability step path and project missing schema when needed.
Direction matters. Use input/state -> local for step inputs and local ->
state/output for step outputs. Run `wf draft validate <workspace_id>` after
this command.
state/output for step outputs. If the workflow schema field already exists,
the command reuses it and updates the step binding. For pure input-map edits
where schema is already known, `wf draft set-input --merge` is also valid.
Run `wf draft validate <workspace_id>` after this command.
"""
context = load_cli_context(ctx)
emit_json(